Utilizing Text Mining to Extract Critical Indicators for Wetland Health Evaluation

Wetland is one of the important ecological systems of the earth, it boasts necessary functions, such as, regulating climate, water supply, purifying water, protecting biodiversity and others. Nowadays, wetland health assessment has become a major direction in wetland research, so the optimization of wetland evaluation is significant for sustainable global development. However, most wetland assessments don’t have agreed indicators and criteria. So, this study aims to use data mining technology to explore the major factors related to relationships about evaluation. Web of Science, EI, Scopus are used to search for training samples, the topic word was “wetland health”, finally 100 articles were selected. These articles compose a corpus which is the basis for text mining. The study mains to apply the online data and web-based text mining technique is proposed to evaluate wetland health status and explore new ideas for wetland health evaluation. Results show that “groundw”, “people”, soil, policy and so on are important and can be called key words. Then, these key words can be used as indicators and they can provide data to the further study.